JDSM PRO/CON DEBATE http://dx.doi.org/10.15331/jdsm.5366 PRO: Upper Airway Resistance Syndrome Represents a Distinct Entity from Obstructive Sleep Apnea Syndrome Lauren Tobias, MD; Christine Won, MD, MS Yale Centers for Sleep Medicine, Yale School of Medicine, New Haven, CT

Controversy exists over whether the upper airway resistance widely across the United States. We agree that sufficient data syndrome (UARS) represents an entity whose pathophysiology exists to accept NCPTs used in combination with respira- and clinical characteristics are distinct from those of obstruc- tory inductive plethysmography (RIP) volume signals as an tive sleep apnea syndrome (OSAS). Many clinicians remain adequate substitute for esophageal pressure monitoring2,3 unconvinced of its clinical relevance as a unique disorder and and certainly one with greater accuracy than a thermistor, instead believe it lies along the same spectrum as OSAS. We for detecting flow-limited respiration in UARS.4 The superior believe that ample evidence suggests UARS indeed represents tolerance of NCPTs by patients renders it a more convenient a separate clinical phenomenon as opposed to simply a less means of identifying subtle breathing abnormalities during severe form of OSAS, and that there is utility in considering it sleep. The presence of flow limitation on the NCPT appears as as a separate disorder. a flattening of the normal bell-shaped curve of a normal breath UARS is a form of sleep disordered breathing characterized with a drop in the amplitude by < 30% compared to normal by repeated increases in upper airway resistance with concomi- breaths immediately preceding the drop. tant increased respiratory effort, resulting in brief arousals. Based on this definition alone, UARS would seem to poten- Such events are termed respiratory effort-related arousals or tially represent a milder degree of upper airway obstruction RERAs, where arousals are defined as a brief shift in alpha or than is present in OSAS. If UARS existed simply on a spec- fast theta frequency on the electroencephalogram (EEG) lasting trum with OSAS, however, we would expect to see the symp- from 3–10 seconds. RERAs are distinct from apneas or hypop- toms of this disorder on a continuum as well. Instead, research neas in that they lack frank apneas or oxygen desaturation and supports several symptoms unique/distinct to UARS and less are typically shorter (1 to 3 breaths), thereby failing to meet predominant in OSAS (Table 1). the generally accepted criteria for either apneas or hypopneas. One of the central arguments in favor of UARS as a distinct RERAs are distinct from apneas or hypopneas in that they lack entity stems from the differences in the population it affects. frank cessation of airflow or oxygen desaturation and are typi- UARS patients tend to be leaner, with a mean BMI ≤ 25 kg/m2, cally shorter (one to three breaths). By definition, the apnea- younger, with a mean age of 37.5 years, and equally present in hypopnea index (AHI) in patients with UARS is less than five. males and females5 (although representing a greater propor- The gold standard measurement of RERAs in UARS is tion of sleep related breathing disorders in women6). Cepha- considered esophageal pressure monitoring (Pes), which detects lometry has revealed craniofacial abnormalities in the upper progressive elevations in intrathoracic pressures with respira- airway anatomies of many UARS patients, including the pres- tion leading up to an arousal. Most early publications of UARS ence of a long face, short and narrow chin with reduced mouth utilized esophageal pressure monitoring and several abnormal opening, retrolingual narrowing, increased overjet, high and 1 7 forms of Pes tracings were described. Pes crescendo is charac- narrow hard palate. terized by a progressively increased negative peak inspiratory A second argument in support of UARS as a distinct entity pressure in each breath, terminating with either an arousal draws its support from the presenting complaints of these or burst of delta wave on EEG. A second abnormality seen patients. UARS may occur in the absence of clinically signifi- involves sustained continuous respiratory effort with a rela- cant snoring and may be an occult cause of excessive daytime tively stable and persistent negative peak inspiratory pressure sleepiness.8 UARS patients generally present with more subjec- seen on the Pes tracing to a degree greater than seen in baseline, tive perception of daytime dysfunction in association with 9 non-obstructed breaths. The third form is esP reversal, charac- sleepiness than do OSAS patients. They complain of worse terized by a sequence of increased respiratory efforts followed subjective sleep quality than OSAS patients as measured by by a sudden decrement in respiratory effort indicated by a less standardized scales of insomnia, subjective sleepiness and negative peak inspiratory pressure. sleep quality10 and higher rates of insomnia related to sleep Despite its utility, esophageal pressure monitoring has not initiation.6 Somatic complaints are more common and distinct been routinely adopted as part of standard polysomnographic personality characteristics are seen in these patients. Patients setup in most sleep laboratories, since it involves the semi-inva- with UARS have been noted to complain more frequently of sive placement of a pediatric feeding catheter into the patient’s chronic insomnia and daytime sleepiness or fatigue than those nostril down to the esophagus and the potential discomfort with OSAS. Reports of headaches, vasomotor rhinitis, irritable associated with this procedure. Ample evidence now suggests bowel syndrome, difficulty in concentrating, and depressed that RERAs may be adequately detected with nasal cannula mood have been described in association with UARS more pressure transducers (NCPTs) and this technology has been frequently than OSAS. Patients with UARS tend to score more

Journal of Dental Sleep Medicine 21 Vol. 3, No. 1, 2016 Pro/Con Debate—Tobias and Won

Table 1—Clinical Features in upper airway resistance syndrome (UARS) and obstructive sleep apnea syndrome (OSAS). UARS OSAS Epidemiology Children Age All ages (mean age 38 years old) Males > 40 years old Females after menopause Male-to-female ratio 1:1 2:1 Body habitus Lean or normal (BMI often < 25 kg/m2) Often overweight or obese Blood pressure Low or normal High Neck circumference Low or normal Large Clinical Presentation Snoring Common but may be minimal or absent in 10–15% Near-universal Witnessed apneas Absent Common Excessive daytime sleepiness, fatigue, morning headaches, Excessive daytime sleepiness, Daytime Symptoms myalgias, difficulty concentrating morning headaches Frequent nocturia, difficulties initiating sleep, insomnia, Snoring, gasping, witnessed Sleep Disturbances bruxism, restless legs syndrome, unrefreshing sleep apneas, nocturia Autonomic Nervous System Hypotension, orthostasis, cold hands and feet Rare Functional somatic syndrome Depression, anxiety, chronic fatigue syndrome, irritable bowel Rare associations syndrome, fibromyalgia Polysomnography Sleep onset latency Long Short AHI < 5/h ≥ 5/h

Minimum O2 saturation > 92% Often < 92% Respiratory effort-related arousals Predominant Minimal Cyclic alternating patterns Frequent Less common Power spectral EEG analysis Higher α power, higher δ in stage REM Less α or δ

Adapted from Bao and Guilleminault, Table 1 and Table 2.1 AHI, apnea-hypopnea index as events per hour. strongly toward neuroticism than OSAS patients on person- In patients with UARS, CAP is a marker of sleep instability and ality inventories8 and demonstrate increased somatic arousal poor sleep quality, and correlated with subjective symptoms of as measured by self-report questionnaires.11 These observations sleepiness and fatigue. Sleep disturbances in this population have led to the suggestion that UARS may represent a func- are often identifiable only with sensitive measures such as CAP tional somatic syndrome such as chronic fatigue syndrome and analysis and not with traditional diagnostic scoring systems.14 fibromyalgia. There are also objective measures suggesting that Fourth, one of the pathologic lesions present in OSAS— UARS is distinct from OSAS. Patients with UARS have been local neurogenic lesions in the pharynx and upper larynx that noted to perform more poorly than OSAS patients on tests of interfere with maintenance of normal airway patency—does psychomotor vigilance,12 a proxy for daytime attentional func- not appear to be present in patients with UARS.1 It has been tion. In retrospective study of patients at an academic sleep hypothesized that this preservation of normal sensory input center, a model fit to predict hypersomnolence among patients from the upper airways leads to faster arousal and recovery with both OSAS and UARS significantly underestimated of normal breathing prior to the point of reaching levels of hypersomnolence in UARS patients.13 hypoxemia seen in OSAS. A third line of argument comes from electroencephalo- A key question in this discussion is whether our current graphic spectral analysis, where EEG power is characterized for definition of OSAS, which allows for the scoring of hypop- each sleep epoch on polysomnography. Patients with UARS are neas (reduced airflow or flow limitation) terminating not only noted to have a general increase in alpha rhythm and relatively in desaturation, but also in arousals, sufficiently captures all more delta power noted during stage REM sleep, in contrast to of those patients with clinical symptoms resulting from said the reductions in both of these frequencies commonly observed arousals. We argue that it likely does not. OSAS is character- in patients with OSAS.1 Furthermore, the presence of cyclic ized by multiple pathologic perturbations including hypox- alternating pattern (CAP) on EEG has also been observed with emia and re-oxygenation, increased intra-thoracic pressure higher frequency in patients with UARS than those with OSAS. and mechanical load, sympathetic activation, elevations of

Journal of Dental Sleep Medicine 22 Vol. 3, No. 1, 2016 Pro/Con Debate—Tobias and Won inflammatory markers, and arousals. The degree to which of influenced the description of this entity. Specifically, most of each of these pathologic events contributes to adverse clinical the initial investigations by Guilleminault and colleagues outcomes likely differs and growing research is beginning to occurred in a population of patients who had presented to elucidate these differences. Indeed, evidence suggests that the sleep clinics. It stands to reason that such patients differ from arousals and associated sleep disruption which are central to the population of patients at large, and even from those who the pathophysiology of UARS may be sufficient in themselves might have been randomly recruited from primary care to cause adverse outcomes, even in the absence of hypox- clinics, in that they suffered symptoms stereotypical enough emia. Human studies have reported associations between to result in subspecialty referral. This issue of sample bias may arousals and subjective sleepiness, changes in hormone secre- be even more relevant to sleep disorders than other medical tion patterns, increased metabolic rate and increased sensory disorders, since detailed sleep histories are not routinely elic- arousal threshold.15 Several studies have demonstrated a ited in primary care settings.19 Without comparison to the positive association between the number of arousals and general population, it is therefore impossible to estimate the awakenings seen on polysomnogram (PSG) and presence true risk conferred by polysomnographic findings of UARS as of hypertension. Evidence suggests that brief arousals from they relate to specific complaints or clinical outcomes. The first sleep, even for a single night, may affect levels of sleepiness the step to classification of any disease requires its recognition in a following day. For example, Philip et al. produced nocturnal narrow patient population, but future work should more rigor- auditory stimuli to elicit arousals and demonstrated a signifi- ously evaluate the prevalence of UARS symptoms and PSG cant reduction in mean sleep latency on next-day multiple findings in all patients, not simply in those with easily identifi- sleep latency testing (MSLT).16 In a similar experiment, Martin able symptoms that drive them to seek care at a sleep clinic. and colleagues demonstrated effects of such stimulation on Even if the above definition of UARS evolves over time, as mood and cognitive function.17 occurs with many disorders, it remains useful as a starting Most initial descriptions of disease began with clinical point. Indeed, our understanding of the pathophysiology of observations of the most pronounced, easily identifiable OSAS continues to evolve, with recent work suggesting the examples of a disorder. Over time, as the less “classic” mani- existence of multiple phenotypes of OSAS, characterized by festations of a disease are characterized and its prevalence the presence or absence of hypoxemia,20 genioglossus muscle explored in other populations, it is often recognized that the responsiveness during sleep, arousal threshold, and presence initial ontology was oversimplified. A disease’s expression of loop gain.21 A recent application of cluster analysis from the in some populations may not be recognized until later on. Icelandic Sleep Apnea Cohort has yielded a further classifica- This was in fact the case with obstructive sleep apnea, where tion scheme based on clinical subtypes that may eventually reports of fatigue, headache, and mood disturbance rather help us to identify patients with sleep apnea based on predomi- than “classic” symptoms of snoring and witnessed apneas,18 nant presenting symptoms.22 We anticipate that these lines were observed more commonly in women, and were not recog- of inquiry will be further pursued in coming years, leading nized for several years after its initial description. Once this to a refined understanding and revised definitions of the realization occurred, OSAS was recognized to be more widely sleep-related breathing disorders, their variable clinical and prevalent in women than was previously believed. Particularly pathophysiologic profiles, and in turn, to modified treatment in the case of UARS, we believe that erring in the direction pathways for managing these patients. The true prevalence of of over- as opposed to under-diagnosis is prudent. The former UARS in the general population is not known, but we suspect it risks needlessly treating some additional patients with a virtu- is substantial, and therefore feel it warrants further study. ally harmless therapy that may be withdrawn at any point; the In summary, we feel that UARS indeed represents a unique latter risks missing an opportunity to improve patients’ quality sleep-related breathing disorder distinct from OSAS, the of life significantly. If we fail to acknowledge that UARS may heterogenous nature of which has only recently begun to be represent a distinct clinical syndrome, we may less vigilantly elucidated. If the disorder truly existed along a continuum ensure the scoring of RERAs and therefore miss opportunities with OSAS, we would expect to see a dose-dependent response to correlate these events with clinical outcomes in the future.6 between the degree of sleep disordered breathing and clin- We acknowledge that there are limitations in the current ical symptomatology. As the aforementioned studies make research on UARS, but view these as constructive starting clear, such a relationship has not been consistently seen. points for further investigations as opposed to justification to Rather, several symptoms appear to occur with increased dismiss the entity altogether. As the Greek playwright Sopho- frequency among patients with UARS and not in those with cles stated, “Look and you will find it—what is unsought will OSAS. Further work is needed in order to advance our under- go undetected.” One limitation of research to date on UARS standing of UARS and its relationship to the more convention- involves the absence of a standardized definition across ally accepted sleep breathing disorders so that symptomatic research groups and even within the same groups over time, as patients do not continue to go untreated. well as heterogeneity in the characterization of events that have been considered to meet criteria as RERAs. A second limita- CITATION tion is that much of the research on UARS has been conducted by only a few research groups and not yet replicated by other Tobias L, Won C. PRO: Upper airway resistance syndrome teams in different patient populations. Finally, a third limita- represents a distinct entity from obstructive sleep apnea tion is the potential of sample bias in existing literature to have syndrome. Journal of Dental Sleep Medicine 2016;3(1):21–24.

Journal of Dental Sleep Medicine 23 Vol. 3, No. 1, 2016 Pro/Con Debate—Tobias and Won

REFERENCES 15. Jaimchariyatam N, Rodriguez CL, Budur K. Sleep-related cortical arousals in adult subjects with negative polysomnography. Sleep 1. Bao G, Guilleminault C. Upper airway resistance syndrome--one Breath 2014;19:989–96. decade later. Curr Opin Pulm Med 2004;10:461–7. 16. Philip P, Stoohs R, Guilleminault C. Sleep fragmentation in normals: a 2. Epstein MD, Chicoine SA, Hanumara RC. Detection of upper airway model for sleepiness associated with upper airway resistance syndrome. resistance syndrome using a nasal cannula/pressure transducer. Chest Sleep 1994;17:242–7. 2000;117:1073–7. 17. Martin SE, Engleman HM, Deary IJ, Douglas NJ. The effect of sleep 3. Ayappa I, Norman RG, Krieger AC, Rosen A, O’Malley RL, Rapoport fragmentation on daytime function. Am J Respir Crit Care Med DM. Non-Invasive detection of respiratory effort-related arousals 1996;153:1328–32. (REras) by a nasal cannula/pressure transducer system. Sleep 18. Redline S, Kump K, Tishler PV, Browner I, Ferrette V. Gender 2000;23:763–71. differences in sleep disordered breathing in a community-based 4. Montserrat JM, Badia JR. Upper airway resistance syndrome. Sleep sample. Am J Respir Crit Care Med 1994;149:722–6. Med Rev 1999;3:5–21. 19. Senthilvel E, Auckley D, Dasarathy J. Evaluation of sleep disorders in 5. Exar EN, Collop NA. The upper airway resistance syndrome. Chest the primary care setting: history taking compared to questionnaires. 1999;115:1127–39. J Clin Sleep Med 2011;7:41–8. 6. Guilleminault C, Kirisoglu C, Poyares D, et al. Upper airway resistance 20. Palma J-A, Iriarte J, Fernandez S, et al. Characterizing the phenotypes syndrome: a long-term outcome study. J Psychiatr Res 2006;40:273–9. of obstructive sleep apnea: clinical, sleep, and autonomic features of 7. Pépin JL, Guillot M, Tamisier R, Lévy P. The upper airway resistance obstructive sleep apnea with and without hypoxia. Clin Neurophysiol syndrome. Respir Int Rev Thorac Dis 2012;83:559–66. 2014;125:1783–91. 8. Kristo DA, Lettieri CJ, Andrada T, Taylor Y, Eliasson AH. Silent 21. Eckert DJ, White DP, Jordan AS, Malhotra A, Wellman A. Defining upper airway resistance syndrome: prevalence in a mixed military phenotypic causes of obstructive sleep apnea. Identification of novel population. Chest 2005;127:1654–7. therapeutic targets. Am J Respir Crit Care Med 2013;188:996–1004. 9. Deary V, Ellis JG, Wilson JA, Coulter C, Barclay NL. Simple snoring: 22. Ye L, Pien GW, Ratcliffe SJ, et al. The different clinical faces of not quite so simple after all? Sleep Med Rev 2014;18:453–62. obstructive sleep apnoea: a cluster analysis. Eur Respir J 2014;44:1600–7. 10. So SJ, Lee HJ, Kang SG, Cho CH, Yoon HK, Kim L. A comparison of personality characteristics and psychiatric symptomatology between SUBMISSION & CORRESPONDENCE upper airway resistance syndrome and obstructive sleep apnea syndrome. Psychiatry Investig 2015;12:183–9. INFORMATION 11. Broderick JE, Gold MS, Amin MM, Gold AR. The association of somatic Submitted for publication October, 2015 arousal with the symptoms of upper airway resistance syndrome. Sleep Accepted for publication October, 2015 Med 2014;15:436–43. Address correspondence to: Christine Won, MD, PO Box 208057, 333 Cedar St, New Haven, CT 06520-8057; Tel: (203) 785-4163; Fax: (203) 12. Stoohs RA, Knaack L, Blum HC, Janicki J, Hohenhorst W. Differences 785-3634; Email: [email protected] in clinical features of upper airway resistance syndrome, primary snoring, and obstructive sleep apnea/hypopnea syndrome. Sleep Med 2008;9:121–8. DISCLOSURE STATEMENT 13. Gold AR, Gold MS, Harris KW, Espeleta VJ, Amin MM, Broderick JE. The authors have indicated no financial conflicts of interest. Hypersomnolence, insomnia and the pathophysiology of upper airway resistance syndrome. Sleep Med 2008;9:675–83. 14. Guilleminault C, Lopes MC, Hagen CC, da Rosa A. The cyclic alternating pattern demonstrates increased sleep instability and correlates with fatigue and sleepiness in adults with upper airway resistance syndrome. Sleep 2007;30:641–7.

Journal of Dental Sleep Medicine 24 Vol. 3, No. 1, 2016